Skip to content

Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.

ci: run-docs-tests-deep now triggers standalone; pin to fixed engine release

Soumya Medapatia5e91d3d97e4
diff --git a/.github/workflows/docs-agent-eval.yml b/.github/workflows/docs-agent-eval.yml
index 2567cd27e..521883092 100644
--- a/.github/workflows/docs-agent-eval.yml
+++ b/.github/workflows/docs-agent-eval.yml
@@ -24,8 +24,11 @@ jobs:
# Run when the PR carries the label — but for `labeled` events, only when
# the label just added is one of ours (any other label addition would
# otherwise cancel-and-restart an in-flight eval; Cursor Bugbot finding).
+ # Either label triggers on its own (run-docs-tests-deep standalone forces
+ # Deep); labeled events still only count when the added label is ours.
if: |
- contains(github.event.pull_request.labels.*.name, 'run-docs-tests') &&
+ (contains(github.event.pull_request.labels.*.name, 'run-docs-tests') ||
+ contains(github.event.pull_request.labels.*.name, 'run-docs-tests-deep')) &&
(github.event.action != 'labeled' || startsWith(github.event.label.name, 'run-docs-tests'))
# Job-level on purpose: a skipped job never enters the concurrency group,
# so unrelated triggers cannot cancel a running eval. New commits still
@@ -37,7 +40,7 @@ jobs:
contents: read
pull-requests: write # the result comment
# Full-SHA pin required by org policy — bump on docs-agent-eval releases.
- uses: ComposioHQ/docs-agent-eval-ci/.github/workflows/eval-pr.yml@f8c0c11175dfac7371860d5d617ff402e0187dc0
+ uses: ComposioHQ/docs-agent-eval-ci/.github/workflows/eval-pr.yml@7d9833c18e0b6e18e822d3933af478261dd6541f
with:
pr_number: ${{ github.event.pull_request.number }}
mode: ${{ contains(github.event.pull_request.labels.*.name, 'run-docs-tests-deep') && 'Deep' || 'Auto' }}